Élie Michel
@elie-michel.bsky.social
Research Scientist at #Adobe. PhD in Computer Graphics. Author of #LearnWebGPU C++. Creative Coding. Indie game. VFX. Opinions are my own. Writes in 🇫🇷 🇺🇸.
https://portfolio.exppad.com
https://twitter.com/exppad
https://portfolio.exppad.com
https://twitter.com/exppad
Note that 'naga' is the equivalent tool developed by Firefox. It can easily be installed using cargo (the rust build manager): github.com/gfx-rs/wgpu/...
June 1, 2025 at 10:44 AM
Note that 'naga' is the equivalent tool developed by Firefox. It can easily be installed using cargo (the rust build manager): github.com/gfx-rs/wgpu/...
'tint' is the shader compiler developed by Chrome to implement #WebGPU. It has a nice command line interface, but so far there is no official build out...
Wait no more! I share here precompiled binaries of tint CLI: github.com/eliemichel/d...
Wait no more! I share here precompiled binaries of tint CLI: github.com/eliemichel/d...
June 1, 2025 at 10:44 AM
'tint' is the shader compiler developed by Chrome to implement #WebGPU. It has a nice command line interface, but so far there is no official build out...
Wait no more! I share here precompiled binaries of tint CLI: github.com/eliemichel/d...
Wait no more! I share here precompiled binaries of tint CLI: github.com/eliemichel/d...
I've just realized something. It makes much more sense to have the "hello triangle" pointing upside down when learning #WebGPU!
👉 The ongoing "Next" rewrite of my guide reached the Hello Triangle chapter 🥳 eliemichel.github.io/LearnWebGPU/...
👉 The ongoing "Next" rewrite of my guide reached the Hello Triangle chapter 🥳 eliemichel.github.io/LearnWebGPU/...
May 29, 2025 at 9:47 AM
I've just realized something. It makes much more sense to have the "hello triangle" pointing upside down when learning #WebGPU!
👉 The ongoing "Next" rewrite of my guide reached the Hello Triangle chapter 🥳 eliemichel.github.io/LearnWebGPU/...
👉 The ongoing "Next" rewrite of my guide reached the Hello Triangle chapter 🥳 eliemichel.github.io/LearnWebGPU/...
🏅Honored to have been awarded at #Eurographics25 for our paper on #LipschitzPruning to speed-up SDF rendering!
👉 The paper's page: wbrbr.org/publications...
Congrats to @wbrbr.bsky.social, M. Sanchez, @axelparis.bsky.social, T. Lambert, @tamyboubekeur.bsky.social, M. Paulin and T. Thonat!
👉 The paper's page: wbrbr.org/publications...
Congrats to @wbrbr.bsky.social, M. Sanchez, @axelparis.bsky.social, T. Lambert, @tamyboubekeur.bsky.social, M. Paulin and T. Thonat!
May 19, 2025 at 9:54 AM
🏅Honored to have been awarded at #Eurographics25 for our paper on #LipschitzPruning to speed-up SDF rendering!
👉 The paper's page: wbrbr.org/publications...
Congrats to @wbrbr.bsky.social, M. Sanchez, @axelparis.bsky.social, T. Lambert, @tamyboubekeur.bsky.social, M. Paulin and T. Thonat!
👉 The paper's page: wbrbr.org/publications...
Congrats to @wbrbr.bsky.social, M. Sanchez, @axelparis.bsky.social, T. Lambert, @tamyboubekeur.bsky.social, M. Paulin and T. Thonat!
New update post about the 🚧 Ongoing work! 🚧 in my LearnWebGPU C++ guide!
On patreon: www.patreon.com/posts/ongoin...
On Discord: discord.gg/2Tar4Kt564
Outline:
🔹 The LearnWebGPU guide
🔹 WebGPU-distribution
🔹 RenderDoc
🔹 WebGPU-C++
🔹 WebGPU spec
🔹 Dawn
🔹 wgpu-native
🔹 GLFW and SDL
🔹 Slang x WebGPU
On patreon: www.patreon.com/posts/ongoin...
On Discord: discord.gg/2Tar4Kt564
Outline:
🔹 The LearnWebGPU guide
🔹 WebGPU-distribution
🔹 RenderDoc
🔹 WebGPU-C++
🔹 WebGPU spec
🔹 Dawn
🔹 wgpu-native
🔹 GLFW and SDL
🔹 Slang x WebGPU
May 10, 2025 at 10:21 PM
New update post about the 🚧 Ongoing work! 🚧 in my LearnWebGPU C++ guide!
On patreon: www.patreon.com/posts/ongoin...
On Discord: discord.gg/2Tar4Kt564
Outline:
🔹 The LearnWebGPU guide
🔹 WebGPU-distribution
🔹 RenderDoc
🔹 WebGPU-C++
🔹 WebGPU spec
🔹 Dawn
🔹 wgpu-native
🔹 GLFW and SDL
🔹 Slang x WebGPU
On patreon: www.patreon.com/posts/ongoin...
On Discord: discord.gg/2Tar4Kt564
Outline:
🔹 The LearnWebGPU guide
🔹 WebGPU-distribution
🔹 RenderDoc
🔹 WebGPU-C++
🔹 WebGPU spec
🔹 Dawn
🔹 wgpu-native
🔹 GLFW and SDL
🔹 Slang x WebGPU
Starting to track down the usage of #WebGPU resources during a frame in my custom #RenderDoc driver!
(Don't mind the usage field, it's a placeholder value for now)
(Don't mind the usage field, it's a placeholder value for now)
May 4, 2025 at 10:43 PM
Starting to track down the usage of #WebGPU resources during a frame in my custom #RenderDoc driver!
(Don't mind the usage field, it's a placeholder value for now)
(Don't mind the usage field, it's a placeholder value for now)
Starting to nest events in the #WebGPU driver for #RenderDoc, how do you think I should handle these "WriteBuffer" that occur while encoding a "RenderPass"?
Because chronologically they are submitted before the render pass even though the API call occurs after.
Because chronologically they are submitted before the render pass even though the API call occurs after.
May 2, 2025 at 10:53 AM
Starting to nest events in the #WebGPU driver for #RenderDoc, how do you think I should handle these "WriteBuffer" that occur while encoding a "RenderPass"?
Because chronologically they are submitted before the render pass even though the API call occurs after.
Because chronologically they are submitted before the render pass even though the API call occurs after.
Close-up on the captured WebGPU API calls
April 30, 2025 at 7:41 AM
Close-up on the captured WebGPU API calls
Who would be interested in a version of #RenderDoc that captures and replays calls to the #WebGPU API (rather than calls to the underlying DirectX/Vulkan/Metal API)?
This is an early test that only lists the API calls, but already promissing! Will share when usable.
This is an early test that only lists the API calls, but already promissing! Will share when usable.
April 30, 2025 at 7:38 AM
Who would be interested in a version of #RenderDoc that captures and replays calls to the #WebGPU API (rather than calls to the underlying DirectX/Vulkan/Metal API)?
This is an early test that only lists the API calls, but already promissing! Will share when usable.
This is an early test that only lists the API calls, but already promissing! Will share when usable.
This report of Wells Fargo is also very informative/worrysome and points to many interesting sources: www.wellsfargoadvisors.com/research-ana...
April 27, 2025 at 11:24 AM
This report of Wells Fargo is also very informative/worrysome and points to many interesting sources: www.wellsfargoadvisors.com/research-ana...
The most commonly cited source about the growth of energy need for compute is this IEA report: iea.blob.core.windows.net/assets/6b2fd... (screenshot of p31)
April 27, 2025 at 11:22 AM
The most commonly cited source about the growth of energy need for compute is this IEA report: iea.blob.core.windows.net/assets/6b2fd... (screenshot of p31)
The results: Our approach provides more expressive manipulation capabilities than solutions that do not augment the implicit function to help tracking down a point's identity!
More info and results on the web page! 👉 eliemichel.github.io/SdfManipulat... 7/8
More info and results on the web page! 👉 eliemichel.github.io/SdfManipulat... 7/8
December 3, 2024 at 8:33 AM
The results: Our approach provides more expressive manipulation capabilities than solutions that do not augment the implicit function to help tracking down a point's identity!
More info and results on the web page! 👉 eliemichel.github.io/SdfManipulat... 7/8
More info and results on the web page! 👉 eliemichel.github.io/SdfManipulat... 7/8
The solution: We #augment the output of the implicit function so that it also returns a vector that represents the #identity of the point. This then enables using the conveniently called Implicit Function Theorem en.wikipedia.org/wiki/Implici..., giving in the end a simple formula. 6/8
December 3, 2024 at 8:33 AM
The solution: We #augment the output of the implicit function so that it also returns a vector that represents the #identity of the point. This then enables using the conveniently called Implicit Function Theorem en.wikipedia.org/wiki/Implici..., giving in the end a simple formula. 6/8
The challenge: The constrained point (the one we click and drag), a priori only exists in the #current version of the shape: it's only once we can track it down in other #variations that we may define its #derivative wrt. procedural parameters.
Not easy, when the entire shape is only #implicit! 5/8
Not easy, when the entire shape is only #implicit! 5/8
December 3, 2024 at 8:33 AM
The challenge: The constrained point (the one we click and drag), a priori only exists in the #current version of the shape: it's only once we can track it down in other #variations that we may define its #derivative wrt. procedural parameters.
Not easy, when the entire shape is only #implicit! 5/8
Not easy, when the entire shape is only #implicit! 5/8
The inspiration: This is typically a case of #DirectManipulation, like for instance #InverseKinematics (IK).
An IK solver estimates how to retro-engineer parameters to match a target constraint... But it requires the #derivative of constrained point position wrt. the procedural parameters. 4/8
An IK solver estimates how to retro-engineer parameters to match a target constraint... But it requires the #derivative of constrained point position wrt. the procedural parameters. 4/8
December 3, 2024 at 8:33 AM
The inspiration: This is typically a case of #DirectManipulation, like for instance #InverseKinematics (IK).
An IK solver estimates how to retro-engineer parameters to match a target constraint... But it requires the #derivative of constrained point position wrt. the procedural parameters. 4/8
An IK solver estimates how to retro-engineer parameters to match a target constraint... But it requires the #derivative of constrained point position wrt. the procedural parameters. 4/8
The problem: This SDF typically has some #procedural parameters exposed to the user. That's nice to explore #variations of the shape... but we only interact indirectly with the shape, through some sliders.
What if we could #directly pull the (implicit) surface of the SDF to explore variations? 3/8
What if we could #directly pull the (implicit) surface of the SDF to explore variations? 3/8
December 3, 2024 at 8:33 AM
The problem: This SDF typically has some #procedural parameters exposed to the user. That's nice to explore #variations of the shape... but we only interact indirectly with the shape, through some sliders.
What if we could #directly pull the (implicit) surface of the SDF to explore variations? 3/8
What if we could #directly pull the (implicit) surface of the SDF to explore variations? 3/8
The input: a Signed Distance Field (SDF) described by a tree or graph of #primitives and #operators (typically, what people do on #shadertoy).
NB: The #SDF is NOT a #mesh, it's an #implicit representation of the shape!
📚 More about SDFs (the classical link): iquilezles.org/articles/ray... 2/8
NB: The #SDF is NOT a #mesh, it's an #implicit representation of the shape!
📚 More about SDFs (the classical link): iquilezles.org/articles/ray... 2/8
December 3, 2024 at 8:33 AM
The input: a Signed Distance Field (SDF) described by a tree or graph of #primitives and #operators (typically, what people do on #shadertoy).
NB: The #SDF is NOT a #mesh, it's an #implicit representation of the shape!
📚 More about SDFs (the classical link): iquilezles.org/articles/ray... 2/8
NB: The #SDF is NOT a #mesh, it's an #implicit representation of the shape!
📚 More about SDFs (the classical link): iquilezles.org/articles/ray... 2/8
🏆 So honored that our paper received the Honorable Mention #Award at #SIGGRAPHAsia!
I really like this paper BTW, it is about procedural #SignedDistanceFields, a simple idea to #differentiate them and thus provide #DirectManipulation ✨
More in the🧵⤵️ 1/8
I really like this paper BTW, it is about procedural #SignedDistanceFields, a simple idea to #differentiate them and thus provide #DirectManipulation ✨
More in the🧵⤵️ 1/8
December 3, 2024 at 8:33 AM
🏆 So honored that our paper received the Honorable Mention #Award at #SIGGRAPHAsia!
I really like this paper BTW, it is about procedural #SignedDistanceFields, a simple idea to #differentiate them and thus provide #DirectManipulation ✨
More in the🧵⤵️ 1/8
I really like this paper BTW, it is about procedural #SignedDistanceFields, a simple idea to #differentiate them and thus provide #DirectManipulation ✨
More in the🧵⤵️ 1/8
Among its many modern language features, #Slang provides #autodiff (forward and backward), which comes very handy for #machine-learning applications! And also a #reflection API, which we use for code generation. 6/8
November 26, 2024 at 9:27 PM
Among its many modern language features, #Slang provides #autodiff (forward and backward), which comes very handy for #machine-learning applications! And also a #reflection API, which we use for code generation. 6/8
Here is what it looks like on the C++ side: a "SomethingKernel" is auto-generated, with a 'createBindGroup' whose signature directly reflects what the #Slang shader asks for! 2/8
November 26, 2024 at 9:27 PM
Here is what it looks like on the C++ side: a "SomethingKernel" is auto-generated, with a 'createBindGroup' whose signature directly reflects what the #Slang shader asks for! 2/8
Sharing a starter setup for using @shader-slang.bsky.social with #WebGPU in C++ with #CMake, either for native or web targets. 🔹 Uses the #reflection API to auto-generate boilerplate code! ✨
👉 github.com/eliemichel/S...
Very happy to take feedback about what's good and bad in this demo! More⤵️ 1/8
👉 github.com/eliemichel/S...
Very happy to take feedback about what's good and bad in this demo! More⤵️ 1/8
November 26, 2024 at 9:27 PM
Sharing a starter setup for using @shader-slang.bsky.social with #WebGPU in C++ with #CMake, either for native or web targets. 🔹 Uses the #reflection API to auto-generate boilerplate code! ✨
👉 github.com/eliemichel/S...
Very happy to take feedback about what's good and bad in this demo! More⤵️ 1/8
👉 github.com/eliemichel/S...
Very happy to take feedback about what's good and bad in this demo! More⤵️ 1/8
For the sake of completeness, here comes the equivalent plot for the #JavaScript side of the API!
November 16, 2024 at 3:17 PM
For the sake of completeness, here comes the equivalent plot for the #JavaScript side of the API!
Is it me of the activity of the #WebGPU-native spec looks like a #deadline is coming? Just saying...😁
(Also, it makes it hard to follow the pace with the #LearnWebGPU guide, but I'm working on it!)
(Also, it makes it hard to follow the pace with the #LearnWebGPU guide, but I'm working on it!)
November 16, 2024 at 3:17 PM
Is it me of the activity of the #WebGPU-native spec looks like a #deadline is coming? Just saying...😁
(Also, it makes it hard to follow the pace with the #LearnWebGPU guide, but I'm working on it!)
(Also, it makes it hard to follow the pace with the #LearnWebGPU guide, but I'm working on it!)